The Effects of Divorce on Children : A Selected Literature Review



Although the research suggests that children of divorce may experience a variety of problems ranging from psychological disturbances to diminished social relationships, the type, severity and persistence of these problems may be mediated (or moderated) by a number of factors. Some of the factors researchers have identified include: child characteristics, such as gender and age at the time of divorce; family characteristics, such as socio-economic status of the custodial household, race, and childrearing skills; and, situational characteristics, such as parental absence, length of time since marital dissolution, conflict, support systems, divorce proceedings, custody arrangements, remarriage, and environmental changes. These factors are discussed below.

3.1 Child Characteristics

3.1.1 Gender

The findings on gender differences in children’s responses to divorce have been contradictory. Some research points to more adjustment problems for boys in divorcing families than for girls (Guidubaldi & Perry, 1985; Hetherington et al., 1979, 1985; Kaye, 1989; Kurdek, 1987); other research finds more negative effects for girls (Farber et al., 1983; Frost & Pakiz, 1990; Slater, Stewart & Linn, 1983; Wallerstein & Kelly, 1975); and some research has found no differences in the effects of divorce on boys and girls (Kinard & Reinherz, 1984; Mechanic & Hansell, 1989; Rosen, 1979; Zill et al., 1993).

Immediately following divorce, Kaye (1989) found that both boys and girls showed poorer performance on achievement tests compared to children from intact families. However, by the fifth year following divorce, boys’ grades and achievement tests were adversely affected, while girls’ were not. Similarly, Hetherington et al. (1979) found that, immediately following the divorce, boys and girls experienced some disruption in play situations, however, the effects appeared to be more sustained in boys. Wallerstein (1985a), in a ten-year follow-up of children who were pre-schoolers at the time of divorce found that although there were no initial sex differences in the effects of divorce. Eighteen months following the divorce, many of the girls appeared recovered, but boys were significantly more troubled at school, in the playground and at home. Five years after the divorce, these sex differences had again disappeared. Guidubaldi and Perry (1985) found that boys in divorced households exhibited more adverse effects than girls, in terms of inappropriate behaviour, work effort, and happiness. Girls with divorced parents, on the other hand, scored higher in locus of control than their counterparts.

Other studies have found more detrimental effects for girls than boys. Slater et al. (1983) found that adolescent girls from disrupted homes had lower self-esteem and more behaviour problems than adolescent boys in similar homelife situations. Furthermore, while female adolescents from disrupted homes reported higher levels of family conflict than females from intact families, the opposite was true for males. Wallerstein and Kelly (1975) found that, one year following divorce, 63 percent of the girls were in worse psychological condition compared to 27 percent of the boys. Frost and Pakiz (1990) found that girls from recently disrupted households reported truancy in higher proportions than their male counterparts and than children from intact families. They were also significantly more dissatisfied with their social network than girls from intact families.

Finally, some studies have found no differences on various effects of divorce between girls and boys (Kinard & Reinherz, 1984; Mechanic & Hansell, 1989; Rosen, 1979). Frost and Pakiz (1990) found no gender differences for self-reported antisocial behaviour among adolescents from divorced families, although they found gender differences in other areas (such as truancy and social networks).

There have been fewer studies examining differences among adult children of divorce. In a study by Farber et al. (1983), clinical directors of college mental health counselling centres said that female adolescents had more difficulty than males in adapting to divorce. However, in a review of the literature, Amato (in press) found minimal sex differences, although women from divorced families appear to attain lower levels of education than those from intact families. In a meta-analysis of 37 studies which examined the long-term consequences of parental divorce for adult well-being, Amato and Keith (1991a) found no support for the contention that parental divorce has more detrimental consequences for males than females. Finally, in a longitudinal study, Zill et al. (1993) found no evidence to support the hypothesis that young adult males were more likely than girls to be vulnerable to the effects of marital disruption.

A possible reason for the contradictory findings related to gender could be that boys and girls may be affected by divorce in different ways. For instance, Kalter (1987) suggests that disruptions in the father-son relationship are linked to a multitude of development interferences in boys. For girls, on the other hand, the emotional loss of father is seen as rejection. Similarly, Healy et al. (1990) argue that boys and girls show sex-role-typical patterns of distress when they see their fathers more often and more regularly - high self-esteem and more behaviour problems for boys, and low self-esteem and fewer behaviour problems for girls. Amato (in press) suggests that the negative effects on social adjustment may be stronger for boys than girls, but in other areas there are no major differences. Other research suggests that girls may be more affected psychologically (e.g., depression) (Peterson & Zill, 1986). Also, it is possible that behaviour problems commonly seen in boys are the more readily observed behaviours than the types of problems that girls have (self-esteem).

Another possible reason for the differing results among studies could be that boys and girls are affected by different aspects of the divorce process. For instance, although Hetherington et al. (1985) found that divorce had more adverse, long-term effects on boys than girls, they found that girls had more adverse effects as a result of remarriage of the custodial mother.

Finally, the heightened divorce adjustment problems for boys found in some research may be less related to gender per se than to characteristics of the postdivorce household arrangements. For instance, Peterson and Zill (1986) found that children living with parents of the opposite sex were especially prone to problem behaviours. However, other studies (e.g., Buchanan, Maccoby & Dornbusch, 1992; Rosen, 1979) have found no significant differences between sex of custodial parent and child’s adjustment. It has also been argued that the differential impact of divorce on children may be linked to parenting styles - particularly with regard to the issue of discipline. Heath and MacKinnon (1988) found that mothers use different amounts of control for sons than daughters. The use of relaxed control by mothers on boys was a high predictor of the child’s competent social behaviour. Further, custodial fathers and mothers have been found to differ in their parenting style, with fathers much less likely to become involved in coercive exchanges with boys than mothers (Grych & Fincham, 1992). The very small number of father custody families and the very selective nature of this arrangement compared to mother custody families means that these studies must be interpreted with a great deal of caution. Grych and Fincham suggest that the question of whether boys or girls are more adversely affected by divorce is quite complex, and the answer is likely to depend on a host of factors such as the sex of the custodial parent, their parenting style, whether they have remarried, the quality of the parent-child relationship, and the amount of contact with the noncustodial parent.

3.1.2 Age at Divorce

Many studies point to the relevance of age at the time of separation for children’s divorce adjustment. Although early findings suggested that separation from a parent at an early age had more negative effects for children than for older youth, this factor has proven to be more complex than was initially believed. In a ten-year follow-up of pre-school children from divorced families, Wallerstein found the initial response to divorce to be worse for younger children, but in later years they appeared better adjusted than their older counterparts (Wallerstein, 1984). She concluded that those who are very young at marital breakup may be less burdened in the years to come than those who are older. Similarly, Amato (1987) found that the majority of children who were very young at the time of divorce reported that they were not strongly affected by the break-up.

The current thinking appears to be that children at every age are affected by divorce, but in differing ways. For example, Krantz (1988) suggests that early separations may be associated with deficits in social and emotional functioning, but not in intellectual functioning. From an examination of numerous studies, Demo and Acock (1988) argue that young children encounter problems with personal adjustment and peer relations, while adolescents encounter problems with sexual relations and antisocial behaviour. Similarly, Zill et al. (1993) found that youth who experienced a family disruption prior to 6 years of age showed poorer relationships with their fathers than those who experienced disruption later in childhood. Landerkin and Clarke (1990) describe how children’s level of development affects their reactions to divorce, although they acknowledge that there may be overlap. The primary reaction among infants may be regression in developmental attainments (e.g., sleeping, eating, language, independence). For pre-schoolers, difficulties may appear in social relationships and separation anxiety. School age children may react with sadness, somatic complaints (e.g., headaches, stomach-aches) and intense anger towards parents. Adolescents may encounter problems establishing an adult identity, demonstrate anger towards self or others, and experience somatic complaints. Finally, Kalter and Rembar (1981) found marital dissolution which occurred very early in a child’s life (2½ years of age or less) was associated with separation-related difficulties; separation during the oedipal phase (2½-6) caused the greatest effects overall on children; and, for those 6 years of age or older, the results were inconsistent.

3.2 Family Characteristics

3.2.1 Socio-economic Status

Often one of the first impacts that divorce has on a child is a dramatic decline in the standard of living in the custodial household (Bean, Berg & VanHook, 1995; Duncan, 1994; Ross, 1995). Krantz (1988) suggests that children belonging to lower socio-economic groups after divorce experience greater hardships. Do these hardships, however, translate into adjustment problems? Some researchers argue that this decline in socio-economic status is directly linked to a variety of problems experienced by the child, such as psychological maladjustment and behavioural difficulties in school. For instance, Nelson (1990) found that family income, rather than marital status, was associated with mothers’ life strains and children’s self-esteem. In addition, Kalter, Kloner, Schreier and Okla (1989) found a negative relation between socio-economic status and children’s adjustment in postdivorce households. However, they suggest that economic deprivation, along with a number of other factors (e.g., inter-parental hostilities, burden of single parenting) take their toll on custodial mothers, which results in poorer adjustment among children.

With a sample of children entering kindergarten, Guidubaldi and Perry (1984) attempted to examine the relation between single-parent status and children’s development, controlling for socio-economic status. They found an association between socio-economic status of parents and intellectual, academic and personal-social development of children. However, even when socioeconomic status was controlled, children from divorced families entered school with significantly less social and academic competence than those from intact families. This indicates that single- parent status may predict poor academic and social competence in addition to, and independent of, socio-economic status. They argue that socio-economic status has a generalized association with both intellectual and non-intellectual measures, while single-parent status is associated with only non-intellectual variables.

3.2.2 Ethno-cultural Background

Very little research has examined ethno-cultural[4] differences among children of divorce. Although there appear to be vast perceptual differences towards kinship, marriage, and divorce cross-culturally, the majority of studies continue to concentrate on Caucasian, and for the most part middle-class, respondents. The results are then interpreted as an indication of the effects of divorce on all children.

However, some research has addressed how various ethno-cultural groups may respond differently to divorce. For instance, in their 1995 study Durndell, Cameron, Knox and Haag (1995) noted radical differences in attitudes towards divorce between native citizens of Rumania and Scotland. Similarly, Tien (1986) noted differences in attitudes towards divorce among Chinese Americans, Korean Americans, and Anglo-Americans.

Some studies have found Hispanic groups to be more affected by family conflict than non-Hispanic whites, while Asians were more affected by a recent divorce (Bean, 1995; Mechanic & Hansell, 1989; Wong 1995). As part of a national survey which examined the relation between adult depression and childhood separation from a parent (due to death, divorce, out of wedlock, etc.), Amato (1991) found that, although white and African American adults who experienced parental absence scored higher on depression than those raised in intact families, these differences, did not appear for Hispanics. He hypothesized that Hispanics may not experience the same negative effects of parental absence because they receive necessary support from their extended families.

Amato (1991) also found that a great deal of the impact of parental absence was mediated by lowered educational attainment and current marital status for whites and African American females, although not for African American males. Furthermore, in a meta-analysis of 37 studies of adults, Amato and Keith (1991a) found that white adults were affected more negatively by parental divorce than African Americans. Lawson and Thompson (1994, 1996) note that African American males are more likely to turn to family and friends, as well as church and other social activities as coping mechanisms following divorce. Each of these studies hypothesized that this was the case because divorce may only marginally lower the quality of life for African Americans, due to the disadvantages they already have.

Following a review of the research, Amato (in press) concludes that there is too little information to reach any conclusions regarding race/ethnicity for children. For adults, he concludes that African Americans appear to be affected less by parental divorce than whites.

3.2.3 Childrearing

The issue of childrearing can encompass a number of aspects, including the effects of employment by the custodial parent on the child, childrearing skills, and adjustment to the divorce by the custodial parent.

The issue of whether employment by the custodial parent has negative effects on children has not been examined in depth. Although it has been suggested that there may be negative effects on the child due to the sole-custody parent (usually the mother) working, a study conducted by Kinard and Reinherz (1984) did not substantiate this claim. Rather, they found that any negative consequences for children of divorce stem from having unemployed rather than employed mothers. However, other researchers have argued that a change in the employment-status of the custodial parent may affect the child. For instance, Mednick, Baker, Reznick and Hocevar (1990) found that instability in maternal employment was associated with negative effects on children.

In a review of the literature, Grych and Fincham (1992) found that parenting styles and discipline practices are linked to the development of behaviour problems in children. This is often the case because, after divorce, parenting is disrupted and discipline frequently becomes inconsistent, both within and between parents. Heath and MacKinnon (1988) argue that childrearing factors are important predictors of children’s social competence in single-parent households. They found that parental acceptance of children was positively related to children’s social competence, while psychological control was negatively related. Further, although they found that social competence related to firm control for males, but moderate control for females, the results indicated that mothers tended to use more lax control for sons than daughters. They suggest that this may provide an explanation for findings which show boys to be worse off than girls in divorces. Heath and MacKinnon found that mother’s unwillingness to exercise firm control over their sons to be a more important determinant of the child’s social competence than father absence. However, Buchanan et al. (1992) found that children living with their fathers had poorer adjustment as a result of poorer monitoring.

The psychological adjustment of the custodial parent after divorce is emerging as a central factor in determining children’s post-divorce adjustment (Cohen, 1995; Kelly, 1993), although the role of maternal adjustment after divorce has been more often examined than the impact of paternal adjustment on children and no studies have looked at the relative contribution of maternal versus paternal adjustment on children. Nor have there been any studies examining the effect and interaction between both parents’ adjustment, conflict, time with both parents, and residence. Weiss (1979) notes that single parents tend to face the following problems which make effective parenting difficult: they often lack adequate support systems; they may feel overburdened by the demands and responsibilities of making all of the daily household decisions alone; they frequently face task overload; and, they may experience emotional overload because of the need to cope with both their own emotional reactions and those of the children. Therefore, it may be particularly difficult for them to discipline consistently and be responsive to their children’s needs. The better the custodial parent adapts to the adversity of the divorce, the more effective he/she can be at providing care, guidance, and support for the children and the more positively adjusted they will be (Kalter et al., 1989). For instance, Nelson (1990) found children’s self-esteem to be directly related to their mother’s life strains. Further, Mednick et al. (1990) found that lower adolescent academic proficiency was related to mother’s adjustment following the divorce. They suggest that the mother’s adaptation to her own personal situation may have a positive influence on the long-term adaptation of her children. Kelly and Wallerstein (1977) suggest that parents should identify the aspects of their behaviour which produce stress on the child and change them to help reduce the negative effects of divorce. Whatever the initial reaction post-divorce, it is important to note that the psychological functioning of parents after separation and divorce improves significantly over time in both men and women (Kelly, 1990).

[4] The term "ethno-cultural" is used in reference to one’s race or ethnic background, as well as their learned, socially acquired life-styles and traditions (i.e., social customs, morals, beliefs, etc.).